We have Case 2 of the Master theorem because
a = 3
b = 3
f(n) = n = Θ(n^log_3(3)) = Θ(n)
Therefore
T(n) = Θ(n*lg(n))
Of course
lg(n) = log_2(n).
Intuitively this means the cost of T is dominated neither by the cost of recursion nor by the work done in the recursion. This is the same as saying that in the recursion tree, the cost of the nodes at each level is asymptotically the same as the cost of leaves.